Exploring data with the theme builder

Jump to section

Overview of the theme builder

The Theme builder tab enables you to explore relationships between your tagged observations, and then create themes based on these relationships. The data in the theme builder updates automatically whenever you create a session, add an observation or add a tag within that study. However, you’ll get the most out of exploring trends in your data once you’ve tagged a substantial number of observations, so we suggest doing this first.

To get the most use out of the theme builder, it’s best to start by exploring themes that relate to your key research objectives. You may want to look at your data from the perspective of:

  • major issues for each task, or steps in a workflow
  • the most significant observations based on those that have been starred during the session
  • parts of the test where participants ran into the most pain points
  • moments where participants experienced moments of delight
  • any suggestions, requests or ideas your participants had
  • patterns in existing behaviors reported by participants at the start of the session
  • finding quotes to support the outcomes of your study.

When exploring the relationships between different tags, pay attention to tags that have a strong relationship, as well as those that don’t. These can help you validate any assumptions you had prior to your research. For example, you may have had concerns that a new navigation will confuse users and find that, in fact, no one encountered any issues finding what they needed.

When your tag combinations start to reveal interesting patterns in your data, you can save these as a theme. Themes group together observations that relate to a similar topic. You can name your theme and add a description for additional context and help other team members understand what you’ve discovered.

How to use the filters

There are 4 layers of filters you can use to ‘ask questions’ and spot patterns in your observations: session name, segments, tags, and tasks if you’ve set them up. You can use one or more of these types depending on what you are looking to discover.

Sessions lets you exclude entire sessions from the analysis. Got data from a participant that you’re not confident about? Leave it out for now until you can follow up.

Segments makes it simple to include or exclude multiple sessions based on the participant segment you created earlier. Only want to see data from people who matched a specific persona? You can do that here.

Tags help you spot connections across all your included sessions and segments, and is where the real power of Reframer lies. As an example, you might spot lots of observations that are tagged with both ‘frustrated’ and ‘login’, hinting that there’s a potential issue with your login screen.

Tasks lets you see all the observations that were captured during your pre-set usability testing tasks. For example, if you wanted to see how all your participants fared during Task 2, you can filter your observations by Task 2 only.

Select a tag in the Theme Builder to display all related observations on the right. As soon as you select a tag, you can create a theme by clicking ‘Save this theme’ and naming it.

In our example study, we’ve selected the tag ‘frustration’ because it’s one of the most common tags in the study so far. Now, only the 8 observations tagged with ‘frustration’ are shown in the right hand column.

We could create a theme with just this tag at this stage, or we can leave it unnamed for now and explore a few more relationships before committing.

Exploring relationships to create themes

​​The tag you’ve selected will influence part of what the Theme Builder shows you next. The list of tags will still display the number of observations each tag appears on, but it’ll now also display the number of observations each tag has in common with your selected tag.

You can refine your search by combining multiple tags. Once you have existing filters applied, the numbers next to the other tags will now indicate how many observations you might see if you combine them. 

► indicates how many observations this tag may have in common with your existing filters, i.e. narrowing with this tag.

◀ indicates how many additional observations this tag will add to your existing filters, i.e. broadening with this tag.


Interpreting the chord diagram

The chord diagram is a visual way to easily explore the relationships between your tagged observations and find key themes. You can choose to view all your tags, or exclude the ones you may not be interested in exploring right then and there.

The core of your Reframer analysis lies in the theme builder, which allows you to create themes by grouping observations that all relate to a similar topic or finding.

Using the theme builder, you can filter your observations by combinations of sessions and tags. When you select a tag, you’ll see all observations in your study that have been assigned that specific tag. From here, you can see any other tags that were used in combination with the one you selected, and broaden or narrow your selection to reveal relationships between them.

When you narrow your tag selection you can reveal observations that contain the specific tags you selected (for example, ‘T1_Login’ AND ‘delight’).

Exploring relationships between tagged observations to find themes

The chord diagram as a whole. Depending on the number of tags you have, this will be more or less complex than this example.